Title: Detecting and analysing QRS complex using automatic tools

Authors: Hanen Chaouch Jebril; Khaled Ouni; Lotfi Nabli

Addresses: Automatic and Signal Processing and Image Laboratory, National School of Engineering, Monastir, Tunisia ' Automatic and Signal Processing and Image Laboratory, National School of Engineering, Monastir, Tunisia ' Automatic and Signal Processing and Image Laboratory, National School of Engineering, Monastir, Tunisia

Abstract: This paper presents a new method for detecting and analysing complex QRS. This method is divided into two main parts: the detection of the complex by the continuous wavelet transform and its control by the linear principal component analysis. The first step consists in determining the localisations of the waves of the QRS complex by leaning on the properties of the regularity and the detection of the singularities offered by CWT. This step allows for the definition of a multivariable matrix presenting several measurements of the various amplitudes of the waves and the segments of the QRS complex. The second step allows studying the data matrix through the linear PCA so as to define the fault variables by using the principle of the fault detection and the fault localisation of the PCA. This framework is applied to some ECG of the MIT/BIH base and is evaluated by comparing it with the data of this same base.

Keywords: ECG signals; electrocardiograms; QRS complex; CWT; continuous wavelet transform; principal component analysis; PCA; cardiac arrhythmias; fault detection; fault localisation; heart rate.

DOI: 10.1504/IJBET.2016.079142

International Journal of Biomedical Engineering and Technology, 2016 Vol.22 No.2, pp.112 - 125

Received: 31 Aug 2015
Accepted: 22 Dec 2015
Published online: 14 Sep 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article